Optimal Average Success Probabilities of Binary $(n,n-1)$ and $(n,n-2)$ Quantum Random Access Codes

Imagine trying to fit a long, detailed story into a tiny envelope. In the classical world of everyday computers, if you try to squeeze a hundred bits of information into a space meant for only fifty, you inevitably lose half the story. You simply cannot retrieve the original details later. But the quantum world, where information is stored in the delicate states of subatomic particles, plays by different rules. It allows for a strange kind of compression where, although you cannot recover the entire original message, you might still be able to pull out any single piece of it with surprising accuracy. This is the realm of quantum random access codes, a tool that tests the very limits of how much information can be hidden in a small quantum system and how reliably it can be found again. Scientists have long wondered about the absolute best performance possible for these codes: if you compress a string of bits into a slightly smaller quantum state, what is the highest chance you can have of correctly guessing any single bit you ask for?

For years, researchers have built specific examples of these codes that seemed to hit a theoretical ceiling, a limit suggested by a mathematical formula involving the square root of the ratio between the compressed size and the original size. While these examples worked perfectly for certain small cases, no one could prove that this ceiling was a hard wall that could never be broken for larger, more complex scenarios. Specifically, the question remained open for two important families of codes: those that compress a string of bits down to just one bit less than the original, and those that compress it down to two bits less. In these high-stakes situations, where the compression is minimal, the existing constructions by other scientists appeared to be optimal, but without a rigorous proof, the possibility remained that a clever new design could squeeze out even more accuracy.

This paper settles that question once and for all. The authors, working at the University of California, Irvine, have provided a complete mathematical proof that the suspected limit is indeed the absolute maximum. They demonstrated that for the specific cases where the quantum state is only one or two bits smaller than the original data, no code can possibly perform better than the existing designs. Their work confirms that the average success probability for retrieving a random bit is exactly the value predicted by the square-root formula, leaving no room for improvement. This means that the constructions created by other researchers are not just good approximations; they are the best possible solutions nature allows for these specific compression ratios.

To reach this conclusion, the team did not rely on computer simulations or guesswork. Instead, they developed a rigorous logical argument that connects the act of measuring a quantum state to the concept of "pretty good measurement," a standard technique in quantum information theory. They imagined a scenario where a decoder tries to reconstruct the entire original string of bits from the compressed quantum state. By analyzing the average number of errors made in this full reconstruction, they were able to derive a strict upper limit on how well any single bit could be recovered. They proved that if a code were to perform better than the suspected limit, it would violate fundamental rules of quantum mechanics regarding how information is distributed and how errors accumulate.

The proof involved showing that the mathematical structure describing the transition from the original data to the reconstructed data must possess a specific property known as positive semidefiniteness. This property acts like a constraint, ensuring that the probabilities of different outcomes behave in a physically consistent way. By combining this constraint with a known limit on how much information can be retrieved from a quantum system, the authors showed that the error rate for the full string reconstruction must be high enough to force the single-bit success rate down to the predicted limit. In the case where the compressed state is one bit smaller, the math shows that the error rate must be at least a certain threshold, which directly translates to the maximum success probability. When the state is two bits smaller, the logic becomes slightly more intricate, requiring a careful accounting of whether the errors are odd or even in number, but the result is the same: the limit holds firm.

The significance of this finding lies in its precision. It transforms a numerical guess into a proven fact, closing the door on the search for better codes in these specific high-rate scenarios. It tells us that the quantum advantage in these situations is exactly what we thought it was, no more and no less. This clarity is essential for the broader field of quantum communication, where understanding the exact boundaries of what is possible helps engineers design efficient networks and secure cryptographic systems. By proving that the existing designs are optimal, the paper provides a solid foundation for future work, allowing researchers to move forward with the confidence that they are working with the best tools the universe has to offer for these particular tasks. The work stands as a definitive answer to a long-standing question, confirming that the theoretical limits of quantum data compression are as tight as the mathematics suggests.